Intent based control of a robotic device

1. A method performed by a semi-autonomous device, comprising:

identifying a first movement of an operator based on monitoring the operator of the semi-autonomous device, the first movement being performed to instruct the semi-autonomous device to travel in a first direction, the operator being remotely located from the semi-autonomous device;

traveling in the first direction of travel within a current environment of the semi-autonomous device in accordance with identifying the first movement;

identifying, via one or more sensors integrated with the semi-autonomous device, one or more objects in the current environment and limitations of the current environment based on traveling in the first direction of travel;

identifying a second movement of the operator after identifying the first movement, the second movement being unassociated with a specific instruction for the semi-autonomous device to travel in a specific direction that is different than the first direction;

identifying, for the first direction of travel, a potential collision with the one or more objects;

inferring a second direction of travel intended for the semi-autonomous device to avoid the collision with the one or more objects, the second direction of travel being inferred in accordance with the second movement, the one or more objects, and the limitations; and

autonomously adjusting, by the semi-autonomous device, the first direction of travel to the second direction of travel to avoid the collision based on inferring the second direction of travel.

2. The method of claim 1 , further comprising transmitting, to the operator, a feed of a current field of view from a vision sensor of the semi-autonomous device.

3. The method of claim 1 , further comprising receiving, from the operator, confirmation of the second direction of travel, wherein the semi-autonomous device travels in the second direction of travel based on receiving the confirmation.

4. The method of claim 1 , wherein the second direction of travel is further determined based on a learned approach or a combination of the learned approach and a modeled approach.

5. The method of claim 4 , wherein:

the learned approach is based on a prior direction of travel of the semi-autonomous device; and

the modeled approach is based on a set of actions corresponding to an intended direction of travel.

6. The method of claim 1 , wherein the semi-autonomous device is trained to avoid the collision with the one or more objects.

7. The method of claim 1 , wherein the second movement is a veering movement toward a direction corresponding to the second direction of travel.

8. A semi-autonomous device, comprising:

a processor; and

a memory coupled with the processor and storing instructions operable, when executed by the processor, to cause the semi-autonomous device to:

identify a first movement of an operator based on monitoring the operator of the semi-autonomous device, the first movement being performed to instruct the semi-autonomous device to travel in a first direction, the operator being remotely located from the semi-autonomous device;

travel, in the first direction of travel within a current environment of the semi-autonomous device in accordance with identifying the first movement;

identify, via one or more sensors integrated with the semi-autonomous device, one or more objects in the current environment and limitations of the current environment based on traveling in the first direction of travel;

identify a second movement of the operator after identifying the first movement, the second movement being unassociated with a specific instruction for the semi-autonomous device to travel in a specific direction that is different than the first direction;

identify, for the first direction of travel, a potential collision with the one or more objects;

infer a second direction of travel intended for the semi-autonomous device to avoid the collision with the one or more objects, the second direction of travel being inferred in accordance with the second movement, the one or more objects, and the limitations; and

autonomously adjust the first direction of travel to the second direction of travel to avoid the collision based on inferring the second direction of travel.

9. The semi-autonomous device of claim 8 , wherein execution of the instructions further cause the semi-autonomous device to transmit, to the operator, a feed of a current field of view from a vision sensor of the semi-autonomous device.

10. The semi-autonomous device of claim 8 , wherein execution of the instructions further cause the semi-autonomous device to receive, from the operator, confirmation of the second direction of travel, wherein the semi-autonomous device travels in the second direction of travel based on receiving the confirmation.

11. The semi-autonomous device of claim 8 , wherein the second direction of travel is further determined based on a learned approach or a combination of the learned approach and a modeled approach.

12. The semi-autonomous device of claim 11 , wherein:

the learned approach is based on a prior direction of travel of the semi-autonomous device; and

the modeled approach is based on a set of actions corresponding to the second direction of travel.

13. The semi-autonomous device of claim 8 , wherein the semi-autonomous device is trained to avoid the collision with the one or more objects.

14. The semi-autonomous device of claim 8 , wherein the second movement is a veering movement toward a direction corresponding to the second direction of travel.

15. A non-transitory computer-readable medium having program code recorded thereon for controlling a semi-autonomous device, the program code executed by a processor and comprising:

program code to identify a first movement of an operator based on monitoring the operator of the semi-autonomous device, the first movement being performed to instruct the semi-autonomous device to travel in a first direction, the operator being remotely located from the semi-autonomous device;

program code to travel, in the first direction of travel within a current environment of the semi-autonomous device in accordance with identifying the first movement;

program code to identify, via one or more sensors integrated with the semi-autonomous device, one or more objects in the current environment and limitations of the current environment based on traveling in the first direction of travel;

program code to identify a second movement of the operator after identifying the first movement, the second movement being unassociated with a specific instruction for the semi-autonomous device to travel in a specific direction that is different than the first direction;

program code to identify, for the first direction of travel, a potential collision with the one or more objects;

program code to infer a second direction of travel intended for the semi-autonomous device to avoid the collision with the one or more objects, the second direction of travel being inferred in accordance with the second movement, the one or more objects, and the limitations; and

program code to autonomously adjust, by the semi-autonomous device, the first direction of travel to the second direction of travel to avoid the collision based on inferring the second direction of travel.

16. The non-transitory computer-readable medium of claim 15 , wherein the program code further comprises program code to transmit, to the operator, a feed of a current field of view from a vision sensor of the semi-autonomous device.

17. The non-transitory computer-readable medium of claim 15 , wherein the program code further comprises program code to receive, from the operator, confirmation of the second direction of travel, wherein the semi-autonomous device travels in the second direction of travel based on receiving the confirmation.

18. The non-transitory computer-readable medium of claim 15 , wherein the second direction of travel is further determined based on a learned approach or a combination of the learned approach and a modeled approach.

19. The non-transitory computer-readable medium of claim 18 , wherein:

the learned approach is based on a prior direction of travel of the semi-autonomous device; and

the modeled approach is based on a set of actions corresponding to the second direction of travel.

20. The non-transitory computer-readable medium of claim 15 , wherein the second movement is a veering movement toward a direction corresponding to the second direction of travel.